Saved in:
Bibliographic Details
Main Author: Erziev, K. O. T.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.04535
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916780633489408
author Erziev, K. O. T.
author_facet Erziev, K. O. T.
contents We propose that benchmarking LLMs on questions which have no reasonable answer actually isn't as silly as it sounds. We also present a benchmark that allows such testing and a method to modify the existing datasets, and discover that existing models demonstrate a performance far from the perfect on such questions. Our code and data artifacts are available at https://github.com/L3G5/impossible-bench
format Preprint
id arxiv_https___arxiv_org_abs_2506_04535
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BSBench: will your LLM find the largest prime number?
Erziev, K. O. T.
Computation and Language
We propose that benchmarking LLMs on questions which have no reasonable answer actually isn't as silly as it sounds. We also present a benchmark that allows such testing and a method to modify the existing datasets, and discover that existing models demonstrate a performance far from the perfect on such questions. Our code and data artifacts are available at https://github.com/L3G5/impossible-bench
title BSBench: will your LLM find the largest prime number?
topic Computation and Language
url https://arxiv.org/abs/2506.04535